Digital Commons Network^™

Trust, Delegation, And Alignment In Human-Ai Decision Making, Erik O. Kimbrough, Brennan Mcdavid, Diba Vazirian

ESI Working Papers

This paper studies delegation to artificial intelligence in a setting where human principals retain the consequences of delegated choices. Participants wrote prompts instructing ChatGPT-4o mini how to choose on their behalf in three canonical economic domains: risky choice, intertemporal choice, and social allocation. We then elicited the compensation participants required to let the AI’s choices count for payment and compared participants’ own choices to choices generated from their prompts. The design produces two central empirical objects: a revealed measure of reluctance to delegate, captured by willingness to accept compensation for AI delegation, and a behavioral measure of alignment, captured by …

Lindenwood Digest, August 19, 2026, Lindenwood University

Lindenwood Digest

• Event & Date: The Lindenwood University Employee Digest was issued on August 19, 2026, announcing campus events including the Walk of Pride on August 21, 2026, Welcome Week professional development sessions on August 20, 2026, and on-campus biometric screenings scheduled for October 6–8, 2026.

• Geographic Grounding: Documented activities and services occur at James Butler Way near the Memorial Arts Building and Warner Hall, the Library and Academic Resources Center (LARC), Evans Commons, and the Spellmann Center at Lindenwood University in St. Charles, Missouri.

• Key Honorees/Awards: Senior Vice President of University Advancement Mindy Spencer was recognized for leadership during her initial …

Denaturalization's Missing Limit, Cassandra Burke Robertson, Irina D. Manta

NULR Online

No abstract provided.

Uni-Body Steam Recovery Engine For Simultaneous Condensate Recovery, Water Heating And Pump-Less Hydraulic Pumping, Jun Ying Yew

Defensive Publications Series

Steam recovery in industrial processing typically requires separate units for condensate recovery, water heating and fluid pumping, each adding capital cost and mechanical complexity. This disclosure describes a uni-body steam recovery engine (“apparatus”) that integrates these three functions — condensate recovery, water heating and hydraulic pumping — within a single structural housing. Hydraulic pumping is achieved without a dedicated mechanical pump by using the pressure differential of the recovered steam to displace condensate and working liquid through the system. The structural arrangement, custom hardware components and control methods governing the cyclic operation of the apparatus are disclosed herein. A proof-of-concept …

Entity Whitelisting Translated To Query-Column-Level Suppression By Introspecting A Detection Rule's Own Entity Mappings, Alpenshield Gmbh

Defensive Publications Series

When a user whitelists an entity (an IP address, account, host, URL, or file hash) against a specific security detection rule, the platform automatically derives which query columns of that rule carry the entity, by reading the rule's own entity-mapping metadata, and materializes column-level suppression entries that the rule can evaluate at query time. In one embodiment the entries are written to a lookup watchlist in the tenant's own workspace, which detection rules join against. In a further embodiment, a shared suppression function is injected into every detection rule's query at deployment time, with per-rule column bindings derived automatically from …

Detecting Silently Blind Security Detections By Joining Rule-To-Source Dependencies Against Per-Tenant Ingestion, And Change-Impact Regression Watch For Detection Rules, Alpenshield Gmbh

Defensive Publications Series

Two related mechanisms for detection-content health in a multi-tenant security operations platform. First, a detection-blindness detector: a maintained dependency graph mapping each detection rule to the log tables its query reads is joined against per-tenant per-table ingestion volume and baselines, yielding a finding when a rule's required input table has stopped ingesting for a tenant, so the rule is silently blind even though it reports no failure. Second, a change-impact regression watch: a time-boxed watch, opened at a detection-rule change by the change pipeline or by an operator, compares post-change alert volume against the pre-change baseline and issues a verdict …

Cross-Channel Authorization Of Security Mitigations: Voice Escalation Bound To An Authenticated Out-Of-Band Approval Surface, Alpenshield Gmbh

Defensive Publications Series

A security operations platform escalates a high-severity incident by placing an automated voice call to a customer contact, during which the contact may authorize a containment action that is approved not on the phone but on an authenticated out-of-band client (a messaging client the contact is already signed into on their own device). The authorization is bound to the call by a correlation token: a per-call code is generated, spoken by the automated agent and displayed on the approval surface, and the approving identity on the out-of-band client must match the contact resolved from the dialed number. The paused containment …

Schema-Validated Field-Level Data Retention Policy Coupled To A Live Database Schema, Driving Automated Per-Field Erasure, Alpenshield Gmbh

Defensive Publications Series

A data platform carries machine-readable retention policy at the granularity of individual tables and columns: for each column, what is stored, why, for how long, and the terminal action (delete the row, null the field, or overwrite the field with a placeholder where the platform marks the column required and a null write would be rejected), while preserving the audit trail. The policy is embedded in version-controlled schema documentation and validated on every build against a periodically captured snapshot of the live database schema, so a policy referencing a column that does not exist, or nulling a column the platform …

Customer-Tenant-Custody Of Third-Party Credentials For Cross-Tenant Security Automation, With Fail-Closed Capability Signaling, Alpenshield Gmbh

Defensive Publications Series

In a multi-tenant managed security service that executes response actions through third-party tools on behalf of customer tenants, the credential for each third-party tool is never held by the service provider. It resides only in the customer tenant's own secret store. At action time, the provider, using a service identity the customer has explicitly consented to, resolves the customer from a management record, authenticates cross-tenant into the customer's secret store, retrieves the credential, exchanges it with the third-party tool, and never persists it provider-side. Capability is signaled fail-closed: a missing or unresolvable configuration returns a distinct "not configured" status that …

Telemetry-Coverage Probes That Structurally Gate Autonomous Disposition Of Security Incidents, Alpenshield Gmbh

Defensive Publications Series

An autonomous security incident analysis pipeline runs deterministic probes that establish, per customer tenant, which telemetry sources are actually available before and during each analysis, and constrains the pipeline's permissible conclusions using the probe results. Probes include a configuration-presence gate for third-party endpoint tooling, a single-row endpoint-inventory query whose three outcomes disambiguate telemetry-available, licensed-without-devices, and not-licensed, a rolling census of log tables that actually ingested data, and explicit consent-gap detection for API scopes. For incident classes whose disposition depends on endpoint evidence, an unavailable-telemetry result caps the analysis confidence and converts an autonomous benign closure into an escalation. A minimum-coverage …

Governed Per-Tenant Knowledge Store For Ai-Assisted Security Incident Analysis, With Adversarial Verification And Customer Dispute Veto, Alpenshield Gmbh

Defensive Publications Series

A multi-tenant security operations platform maintains, for each customer tenant, a store of discrete environment facts that are compiled into a briefing consumed by an AI analysis pipeline for every security incident. Facts carry provenance classes, lifecycle states, evidence counters, validity windows, and stable natural keys. Facts learned automatically from closed incidents pass a two-stage verification: an adversarial language-model pass checks that every concrete identifier in a candidate fact appears verbatim in the incidents it cites, and a deterministic set check validates the citations themselves. Failing candidates carry a machine-readable quarantine annotation naming each unsupported identifier. Activation requires a minimum …

Gpu Ray Tracing And Di!Erentiable Multi-Physics Innovations, Xavier Pillet

Defensive Publications Series

This document, produced with the assistance of GPT-5.2 Thinking and Gemini 3 Reasoning, is released under the Apache License 2.0. It is a voluntary defensive publication (prior art) and therefore enters the prior art upon release under the applicable patent statutes : EPC Art. 54(2) (European Patent Convention), French IPC Art. L 611-11 (French Intellectual Property Code), 35 U.S.C. §102(a) (United States Patent Act), Chinese Patent Law Art. 22(5) (両匡丢氡儡吡嘡丣刡氢), and Japanese Patent Act Art. 29(1) (闘旧避). Description: defensive publication covering enabling inventions around GPU/RT-core ray tracing, differentiable simulation, experimental validation, interoperability (standards), operational security, sustainability, and cross-domain applications (RF/LiDAR/IR/acoustics/seismology/climate/nuclear/biophotonics/medical), …

Model-Agnostic Search-Augmented Llm Pipelines For Production-Scale Company Data Validation, Enrichment And Normalization, Hp Inc

Defensive Publications Series

This paper presents a vendor-agnostic, search-augmented large language model (LLM) pipeline that validates, enriches, and normalizes company records using grounded information from public web sources. The pipeline addresses enterprise requirements for official names, addresses, tax identifiers, revenue, industry classifications, and related compliance metadata where manual validation, database lookups, and rule-based integration pipelines struggle with incomplete, inconsistent, multilingual, and region-specific inputs. The architecture uses a central control framework to orchestrate native-search and tool-calling web-enabled LLM providers under production constraints, including context limits, tokens-per-minute (TPM), requests-per-minute (RPM), maximum batch size, concurrency, incomplete outputs, safety filtering, and provider-specific response formats. It combines batch-admission …

Resource-Constrained Multiprocessing Packet Pipeline With Shared Configurable Elements, N/A, N/A

Defensive Publications Series

A subsystem for network packet processing uses a pool of shared, configurable processing resources. The architecture includes these shared processing elements, a centralized event processing engine, and a partitionable shared memory. The event processing engine accumulates events from the packet pipeline and forwards them as tasks to the processing elements. These elements, which may be implemented as microcode sequencers, execute writable microcode firmware to perform logical and arithmetic operations and handle interrupts independently of a central processing unit. The processing elements access the shared memory to exchange state and packet data between processing stages. This approach improves resource utilization and …

Method And Apparatus For Reducing Semiconductor Package Warpage During Reflow, N/A, N/A

Defensive Publications Series

The technology generally relates to controlling package warpage during reflow soldering using a temporary support structure. The structure, composed of a glass carrier and a high-coefficient of thermal expansion balance film, is applied to the top surface of a package substrate before the reflow process. As temperature increases during reflow, the high-coefficient of thermal expansion balance film expands or shrinks, inducing a controlled downward deformation that acts as a counter-stress. This action balances the internal stresses that cause package warpage, maintaining a flat profile throughout the reflow cycle. The flat package profile facilitates accurate alignment and bonding of its ball …

Automated Clock Domain Crossing Maximum Delay Constraint Application Using Intermediate Anchor Gates, N/A, N/A

Defensive Publications Series

The technology manages clock domain crossing paths by applying a maximum delay constraint to a designated intermediate anchor gate instead of the final destination flip-flop. The register-transfer level configuration instantiates a specific logic gate, such as a MASK AND gate, as this anchor point and assigns it a standardized name. During compilation, a software script automatically discovers these anchors by name and applies a localized, point-to-point delay constraint directly to the pins of the anchors. This pin-to-pin anchoring decouples the timing constraint from the final sampling logic. This approach results in greater placement flexibility for the destination logic across the …

Neuromorphic Soft Swarm Robotics, Xavier Pillet

Defensive Publications Series

This document, produced with the assistance of GPT-5.2 Thinking and Gemini 3 Reasoning, is released under the Apache License 2.0. It is a voluntary defensive publication (prior art) and therefore enters the prior art upon release under the applicable patent statutes : EPC Art. 54(2) (European Patent Convention), French IPC Art. L 611-11 (French Intellectual Property Code), 35 U.S.C. §102(a) (United States Patent Act), Chinese Patent Law Art. 22(5) (両匡丢氡儡吡嘡丣刡氢), and Japanese Patent Act Art. 29(1) (闘旧避). It discloses enabling systems and methods combining soft actuation (HASEL/EAP), event-based vision (DVS), neuromorphic computing (SNN), and swarm intelligence, covering devices/sensors, closed-loop control, …

Memory-Based Lock-Free Inter-Process Communication, Vlad Popa, Andy Hung, Atneya Nair, Mikhail Naganov, Shunkai Yao, Eric Laurent, Jean-Michel Trivi

Defensive Publications Series

A lock-free inter-process communication (IPC) system can utilize shared memory to broadcast system events from a single writer process to multiple reader processes. The system employs a shared memory circular queue and an atomic sequence counter that increments before and after write operations, allowing readers to determine if an active write is occurring. By calculating an active queue index using the sequence counter, reader processes can safely access previously committed historical state data without blocking or busy-waiting. This non-blocking architecture prevents IPC message queue overflows for frozen or suspended background applications, eliminates priority inversion, and avoids the thundering herd problem …

Adaptive Control Of A Media Environment Using Synchronized Physiological Data, Rohit Jain

Defensive Publications Series

Media environments can operate without awareness of a user's physiological state, and some integration methods may be limited by factors such as high-latency cloud processing or simplistic control actions. A system is described that can establish a low-latency communication channel between one or more wearable devices, such as a smart watch or fitness tracker, and a media playback system, for example, a smart television or streaming device. With appropriate user consent, physiological data, such as heart rate variability and electrodermal activity, can be transmitted from the wearable device and processed locally on the media system. Based on this real-time analysis, …

Context-Aware Personalization, Kate Strudwick, Zoe Schladow

Defensive Publications Series

This disclosure describes an automated process for transparent algorithmic profiling and data shaping that allows an individual to actively curate environmental and physiological data to dynamically guide and personalize machine learning models. By moving beyond rigid history logs or binary input structures, the techniques of the disclosure enable a traceable, user-directed taste profile that maps real-world context onto digital preferences and actively seeds future recommendation behaviors to counteract algorithmic flattening. The disclosed system helps Artificial Intelligence (AI) models, including foundation models, generative AI models, and/or large language models (LLMs), capture and interpret multifaceted human data streams to accurately personalize user …

Semantic Graph Reconciliation Using Functional Fingerprinting, Rishabh Singhal, Vijai Kalyan

Defensive Publications Series

Comparing versions of graph-based workflows can present challenges, as methods relying on volatile identifiers may interpret a node rename as separate deletion and addition operations, while other structural comparison techniques can be computationally expensive. Systems and methods are described for semantic comparison using a multi-pass reconciliation algorithm that employs functional fingerprinting. This process can involve generating a content-based hash value for each node after programmatically excluding non-semantic data, such as unique identifiers and graphical coordinates. By comparing these fingerprints, a system can identify nodes that were renamed. This approach can facilitate a linear-time semantic reconciliation of workflow graphs, providing a …

The Dynamic Schema Generation With Error-Triggered Renegotiation For Agent-Tool Binding, Nicolas Zimmermann, Isaac Leimgruber

Defensive Publications Series

Some methods for binding autonomous agents to external tools can rely on static interface schemas, which may be inefficient, for example, by consuming large context windows, and can be brittle to changes in a tool's application programming interface. A system is described where an agent can transmit a semantic intent describing a desired task outcome. In response, one or more tool providers can dynamically generate a minimal, task-specific interface specification tailored to that intent. The system can also feature an error-triggered renegotiation mechanism, allowing an agent to report an execution failure, which prompts the provider to generate and resubmit a …

System-Level Generation Of Dynamic Information Overlays For Media Content, Harsh Rajotiya, Dev Daftari

Defensive Publications Series

Accessing supplemental information while viewing digital media may involve disengaging from the primary content to use separate applications or devices. This document describes a system level framework for generating dynamic informational overlays that can operate independently of specific media applications. The system may utilize a generative artificial intelligence model to interpret user queries, such as voice commands, and orchestrate the retrieval of data from multiple sources. The model can then dynamically design and specify a custom graphical user interface element, such as a chart or table, which can be rendered as an overlay on a main viewing screen (e.g., a …

Identify User Navigation Patterns And Digital Experience Deficiencies Using Probabilistic Sequence Modeling And Llm-Driven Synthetic Personas, Hp Inc

Defensive Publications Series

Understanding how users navigate complex digital platforms is essential for improving conversion, usability, and overall experience. Raw clickstream data generated by web platforms is highly granular, sparse, and noisy, which makes direct URL-level analysis unreliable and computationally unstable. In addition, traditional analytics approaches rely on aggregate metrics or predefined funnels that fail to capture sequential dependencies and heterogeneous user intent. This disclosure describes a method for identifying navigation inefficiencies, behavioral drop-off patterns, and experience deficiencies by combining probabilistic modeling of user navigation sequences with AI-driven synthetic personas. The method models user sessions as discrete-time Markov processes over a reduced semantic …

Evidence-Based Guidance System For Data Science Teams Using Generative Ai Under Uncertainty, Kristoffer Ahlstrom-Vij

Defensive Publications Series

A method for implementing evidence-based guidance for a data science team using generative artificial intelligence under uncertainty is provided. The method includes reviewing existing evidence on generative AI in knowledge work, including empirical literature on productivity effects, cognition and skill formation, and human factors research on automation complacency and cognitive offloading. The method includes evaluating the existing evidence across parallel considerations and deriving a plurality of principles from the existing evidence. The principles include an accountability principle stating that accountability for work cannot be self-certified, an aggregation principle directing to read more but not think less, a writing reservation principle …

Generative Synthesis Of Context-Aware Media Recaps And Alternative Narratives, Dev Daftari, Akash Verma

Defensive Publications Series

Systems and methods can use generative artificial intelligence to synthesize personalized media content. For example, when a user returns to a series, a system can generate a dynamic recap designed to be spoiler-free based on a user's last-viewed point, for instance, by retrieving relevant plot information from a media knowledge graph. The technology could also generate alternative narrative scenes based on user prompts, for example, by using system guardrails to promote stylistic consistency with the source material. These techniques may provide personalized user interactions, potentially improving viewer re-engagement with serialized content and extending interaction with media franchises within a media …

A Generative System For Real-Time Audio Dubbing With Voice And Emotion Transfer, Dev Daftari, Harsh Rajotiya

Defensive Publications Series

Systems and methods are disclosed for real-time media localization that may address potential scalability and fidelity limitations of some conventional approaches. A technical solution may involve a multimodal generative pipeline, which can be implemented on a hybrid edge-cloud architecture. For example, a system can separate dialogue from background audio on a client device (e.g., a smartphone, smart television, personal computer, wearable device, or augmented reality glasses). The isolated dialogue may then be processed to transcribe speech to text and to extract paralinguistic features, such as vocal timbre and emotional tone. This information may be processed by a large language model …

Dynamically Generating Immersive Environments For Extended Reality From 2d Media, Dev Daftari, Akash Verma

Defensive Publications Series

Watching two-dimensional (2D) media in extended reality can involve static virtual environments that may not be contextually linked to the content being viewed, which can reduce immersion. Systems and techniques are described for dynamically generating a 360-degree virtual environment from a 2D media stream. A system may analyze a buffer of video frames using computer vision methods to extract scene context, including three-dimensional geometry and lighting information. This extracted data can then be used as a seed for a generative artificial intelligence model to perform spatial outpainting, extending the original 2D scene into a volumetric environment. This process can transform …

Natural Language Navigation Of Media Via Generative Ai And A Multimodal Index, Dev Daftari, Akash Verma

Defensive Publications Series

Finding specific moments in media can be challenging, as certain methods like timeline scrubbing or keyword searches may not align with a user's descriptive memory of a scene. A system can facilitate media navigation through natural language queries. For example, a system may create a time-stamped, multimodal index of a media asset by analyzing its visual, auditory, and textual content. When a user provides a descriptive query, a generative artificial intelligence model can semantically evaluate the request against the index to identify a corresponding media segment and its timecode. The system could then provide the timecode to a media player …

Cascaded Liability-To-Feedstock Resource Recovery Architectures: Geothermal-Desalination-Membrane-Mineral Stacks, Cryogenic Storage-Air Separation-Datacenter Cascades, And Parallel-Topology Hybrid Storage, Alexander Smith

Defensive Publications Series

Disclosed are cascaded resource-recovery architectures organized by a common design rule: each processing stage is sequenced so that the waste liability of the preceding stage becomes the feedstock of the following stage. Three families are disclosed in enabling detail.

(A) A geothermal-desalination-membrane-mineral stack in which geothermal heat drives power generation via an Organic Rankine Cycle, waste heat drives or preheats desalination, and desalination brine -- already filtered and cooled by the desalination pretreatment train -- supplies a pressure-retarded osmosis salinity-gradient power stage. Geothermal brine separately supplies mineral recovery including direct lithium extraction, silica, zinc, manganese, barium and strontium sulfate. The …